Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!ginosko!uunet!ncrlnk!ncr-sd!hp-sdd!hp-pcd!hplsla!jima
From: jima@hplsla.HP.COM (Jim Adcock)
Newsgroups: comp.object
Subject: Re: OO Case tools
Message-ID: <9450010@hplsla.HP.COM>
Date: 6 Oct 89 19:48:20 GMT
References: <13307@athertn.Atherton.COM>
Organization: HP Lake Stevens, WA
Lines: 29

>I don't know if this helps, but some (most?) of the best programming
>environments in the world are built around dynamic object-oriented
>languages like Lisp Flavors  and Smalltalk.
>
>Of course, the operative word in the above sentence is 'dynamic'.  The
>dynamic (i.e. interactive) nature of the early object-oriented
>languages contributed to much of their power.  The value of
>a dynamic environment was (sadly) overlooked when people designed
>object-oriented extensions to C and Pascal.

ParcPlace has a Smalltalk-like development environment for C++.  In fact 
they make Smalltalk & C++ versions of the same environment.  The environment
features incremental compilation to allow "interactive" style of development.
Given the number of C interpreters, and the rate of growing interest in 
C++,  I bet a number of companies will have C++ interpreters out soon.

Some language developers concentrate on development environment first, at
the expense of compiler efficiency.  Other language developers concentrate
on compiler efficiency first, at the expense of not having good development
tools.  Eventually if a language really catches on, enough $$$ are generated
that users get good compilers AND good development environments.  Then the
only thing holding a language back are any fundamental flaws in the language's
design.

If stuck with a bad development environment, the best one can do is get a 
very fast machine to do development on.  Ditto if stuck with inefficient 
compilers.  Raw MIPS compensates for a lot of problems.  Given the rate
of development of RISC machines, maybe in a couple years this will all be
a "dont-care."